Options For Repeat Move Camera Work With BM

What options does one have to do repeat move camera work where the camera can be programmed to repeat the moves over and over again with pinpoint accuracy.

Should go without saying but I'm looking in the Indie/Rebel filmmaker budget range (not Hollywood)

Would appreciate data or links to data and tests. First hand experience even better. Please no sarcastic comments such as "get a million dollars" etc... I need to make this work.

I have BMPC and would like to purchase the Ursa Mini 4.6 so the options should be translatable to those cameras, although I'm willing to take a look at options using other cameras if need be.

Most of the options I've found are in the $$ stratosphere range.

Motion control has to have gotten cheaper since 1977, right? What's John Dykstra's number?

Motion control has to have gotten cheaper since 1977, right? What's John Dykstra's number?

Thanks for the help. Sorry for stating I'm not interested in sarcastic comments in my OP.

Imagine missing out on the wealth of information you just shared with me.

Have a lovely day.

As you've certainly searched the internet for not-astronmical-priced motion control solutions and come away with nothing, then the answer is most likely what you and I already know: professional motion control is astronomically priced.

Now, you could go a cheaper and more limiting route with those motorized pan/tilt heads that hipsters use on their little sliders. The Kessler solution starts at $2600 and is extremely limited compared to the abilities of the offspring of the Dykstraflex system. There are cheaper variations that do even less.

So, no, proper motion control is ridiculously complex and astronomically expensive.

So, no, proper motion control is ridiculously complex and astronomically expensive.

Define "proper" & "complex.

I understand that you won't get the same flexibility that you would with a Bolt grade system but you can do a lot if you have some basic tools and put in the work.

I've looked at Kessler and I like it. I can't seem to find any in depth look at it in video format but thanks for giving me a heads up.

I just need to see how it would work with a BM camera. It seems both flexible and yet inflexible at the same time from this

.

And here's another.

This seems like quite a good solution on the surface. It can't do what a Bolt does but still, it seems sturdy and accurate... more so than Kessler. At 11k it doesn't break the bank.

Would love to hear from anyone that used any of these - or maybe something else.

Is that a BM camera at 1:05 in the video below ??

Gotta love people that find solutions instead of being bogged down in 'it can't be done' mentality.




Thanks.

We got an Edelkrone setup for our BMPCC and future Mini 4.6.

https://www.edelkrone.com/us/p/138/slider-plus

Is this cheap enough and capable enough for you. Not exactly portable.

http://www.ebay.com/itm/Fanuc-S-420F-In ... SwwE5WVh~7

Is this cheap enough and capable enough for you. Not exactly portable.

Read the fine print. It costs $11K to rig and ship it.

repeat the moves over and over again with pinpoint accuracy

That's going to cost you some money.

The solutions you've been looking at, like Kessler, do not have pinpoint accuracy, or even sufficient precision to be used for compositing. Typically, moco system with that kind of accuracy either use heavy robotics as a base, or they use geared transport to ensure registration. Kessler and the others use ungeared wheels which maintain position by friction. Also, they are lightweight, so the vibration from a passing truck will change the registration.

You might have better luck on cinematography.com or similar sites, and look for a 30-year old Trondeau moco dolly or similar old tech.

Good luck.

I just need to see how it would work with a BM camera.

Well, a camera is a camera. With the timelapse features of the BM cameras, you're safe there in that ballpark, so then you need to worry about how much weight the motion control system can take. If you're using the Pocket or a Micro camera then weight isn't much of an issue probably. If you're just interested in doing timelapses of stars and clouds moving super fast across the sky in a slow slider move, then something like the Kessler option is all you might need.

Again, unless you're doing Star Wars where you need precise, repeatable, programmable camera moves that includes articulation of zoom and focus controls, shutter and aperture like for shooting miniatures and making them look like huge spaceships....proper motion control...then there is no such thing as an affordable option.

So shooting like this:



is different than shooting like this:



going from one extreme to the other.

Here is CamBlock Adventure:



only $11,500

Mark Roberts Motion Control has a wide selection of rigs for various uses and budgets here: http://www.mrmoco.com/cranes-rigs/products/rigs/

So proper motion control -- for, like, visual effects where camera moves need a bunch of elements comped together to make a shot -- is all very expensive and complicated. And unless you're looking to do a lot of it as a profession, then I suggest hiring a VFX company, or renting (or borrowing) something in the medium range like the CamBlock, or buying something smaller like the Kessler, or figure out a different way to shoot what you want, or scrap the whole idea because it's outside your budget.

So, if you just want to shoot timelapse sky videos, go buy any moco rig that supports the weight of the camera and lens you want to use. If you want to comp VFX, then hire someone or lock the camera, shoot in UHD, do the comp and wiggle the camera in post. It just depends on what the final product is intended to be. Good luck.

Jason, you are a patient man. Seems like good advice, but would be good to know the budget range for purchasing the 'pinpoint' accurate solution and the frequency of need that might justify purchase versus equipment rental or VFX services hire.


Sent from my iPad using Tapatalk

repeat the moves over and over again with pinpoint accuracy

That's going to cost you some money.

The solutions you've been looking at, like Kessler, do not have pinpoint accuracy, or even sufficient precision to be used for compositing. Typically, moco system with that kind of accuracy either use heavy robotics as a base, or they use geared transport to ensure registration. Kessler and the others use ungeared wheels which maintain position by friction. Also, they are lightweight, so the vibration from a passing truck will change the registration.

You might have better luck on cinematography.com or similar sites, and look for a 30-year old Trondeau moco dolly or similar old tech.

Good luck.

Thanks - That's an interesting point of view.

I come from the CG world so I'm more than a bit familiar with a lot of of the subject. I've personally done comps from footage done with truly bare bones solutions such as Kessler (motorized sliders). It's true you have more things to worry about and you're limited. But you compensate for that by doing safety takes and overshooting in resolution. It takes more time to think things through and plan things than just setting up a Bolt and running (although that can take long too sometimes) but you can get stellar results so long as you understand what your limitations are. For example: a zoom + trucking + pedestal is out of the question. But any of the individual moves are more than fair game and can come out professional.

Don't know what's in the Kessler modules on the inside but looking at the Kessler photo I can see geared wheels. Matter of fact I don't know of any mechanical system that repeats movements like these with precision without geared wheels. Not sure pneumatics/hydraulics could yield something like that.

30 year old systems are laughable by today's standards. Have you even comped any data off those systems? You just pick up any 1985 movie and watch the desync of the plates.

I find it interesting that cinematographers will spend time and brain power fiddling with their lenses and equipment and end up (if they're skilled) getting Hollywood million dollar like look on a budget, but when it comes to another problem that's more out of their field they throw their hands up in the air and say "Well, you gotta have a million bucks to do that."

Jason, you are a patient man. Seems like good advice, but would be good to know the budget range for purchasing the 'pinpoint' accurate solution and the frequency of need that might justify purchase versus equipment rental or VFX services hire.


Sent from my iPad using Tapatalk

Rick - could go to about 15-20k pending on options. Sorry for not stating this earlier. Thanks

Thanks, Johnny. And is this something where you will want a one-time purchase because you intend to use the equipment on many projects or can your current project support that cost as a service or rental as Jason was suggesting as possible options to get the best results?


Sent from my iPad using Tapatalk

Thanks, Johnny. And is this something where you will want a one-time purchase because you intend to use the equipment on many projects or can your current project support that cost as a service or rental as Jason was suggesting as possible options to get the best results?
Sent from my iPad using Tapatalk

I'm looking to buy rather than rent.

Seeking to add a double take to simple moves like small dollying or panning etc... I'm fully aware that for this price I can't get something that does fast moves on all axis or that sends 3d data to apps like Maya.

I'm just looking for the best bang for the buck which is defined as the ability to redo the move as precisely as possible so I can comp them together in Nuke.

FYI,

http://qz.com/132674/how-robot-movie-ca ... -possible/

Bot & Dolly Industrial Robots...

https://www.facebook.com/335395489836739/videos/383635998346021/

.. but when it comes to another problem that's more out of their field they throw their hands up in the air and say "Well, you gotta have a million bucks to do that."

Not true for all of us
I recently ran into the same situation for a longer term project which finally "forced" me to take action.. currently brewing something together... I'll post more about it once it's out of the early stages.
(not based on industrial robots, more about using things that a lot of us have already but automating/controling them)

Hi Johnny,

What about these guys?

http://www.mrmoco.com

Edit: Sorry, realised you mentioned the Bolt. Just discovered that these guys do the Bolt robot.

Ignore me. Lol

Bot & Dolly Industrial Robots...

https://www.facebook.com/335395489836739/videos/383635998346021/

Yep, this is the way it will go. However, things will get less and less pricey as robotics advances and becomes ubiquitous. There's more than one way to skin a cat of course. Many robotics have capabilities of repeated precise movements, although they are not designed specifically for the filming industry.

.. but when it comes to another problem that's more out of their field they throw their hands up in the air and say "Well, you gotta have a million bucks to do that."

Not true for all of us
I recently ran into the same situation for a longer term project which finally "forced" me to take action.. currently brewing something together... I'll post more about it once it's out of the early stages.
(not based on industrial robots, more about using things that a lot of us have already but automating/controling them)

Glad to hear it. It's inspiring to hear of people that are willing to search for solutions instead of giving in

Keep us posted.

I'll update this thread too with anything I'll be doing.

Maybe it will help the next guy coming along and facing this kind of challenge.

Bot & Dolly Industrial Robots...

https://www.facebook.com/335395489836739/videos/383635998346021/

Yep, this is the way it will go. However, things will get less and less pricey as robotics advances and becomes ubiquitous. There's more than one way to skin a cat of course. Many robotics have capabilities of repeated precise movements, although they are not designed specifically for the filming industry.

The general trend in robotics is to make them more general purpose, easier to program and less expensive. Couple that with the fact that robots have been in wide use for over a couple of decades and as time goes on, lots of relatively inexpensive used robots will come on the market. The Chinese have also jumped into manufacturing robots, so I expect to see inexpensive and capable new robots as well in the near future.

The interesting thing about these robots is that their lift capacity allows them to not only move cameras, but large props and actors.

Lots of new ways to get the shot.

Bot & Dolly Industrial Robots...

https://www.facebook.com/335395489836739/videos/383635998346021/

Yep, this is the way it will go. However, things will get less and less pricey as robotics advances and becomes ubiquitous. There's more than one way to skin a cat of course. Many robotics have capabilities of repeated precise movements, although they are not designed specifically for the filming industry.

The general trend in robotics is to make them more general purpose, easier to program and less expensive. Couple that with the fact that robots have been in wide use for over a couple of decades and as time goes on, lots of relatively inexpensive used robots will come on the market. The Chinese have also jumped into manufacturing robots, so I expect to see inexpensive and capable new robots as well in the near future.

The interesting thing about these robots is that their lift capacity allows them to not only move cameras, but large props and actors.

Lots of new ways to get the shot.

You're absolutely right. Robotics will follow the same path as computing where you have the computer as a platform on which you load your programs for different things that vary widely (eg: from databases, to word processing to CG etc)

A quick and dirty method is to remember that basically anything that can do repeated movements and can have a camera slapped on it is fair game.

For example, I've tried using a display platform that spins (very slowly) for pans where you need 2 instances. You have to manually sync the footage in the comp which can be tricky but if you shoot safety takes and have a high resolution (compared to the output) you can get good results. You just make an animated matte between the plates and pick a non-straight line (because it's harder to notice & more forgiving) you should have some great results.

For a shot that lasts 1 or 2 sec on the screen it works.

A quick note on that: the camera has to be static so it works better if you don't press the rec button from the camera itself (thereby moving it from position by the simple act of touching it)

Personally though I'm leaning towards camblock as best in my price range 15-20k.

Somebody's working on a small 3D printed moco robot that works with a goPro. Can't find the link right now, but it's out there.

Should be getting one of these any day now.

http://www.ebay.com/itm/111704888041

Couldn't see anyplace that specified what it can lift.

Imagine putting a set around the person with a camera facing them for a very realistic crash scene they get to walk away from.

Should be getting one of these any day now.

http://www.ebay.com/itm/111704888041

Couldn't see anyplace that specified what it can lift.

This has a 7 inch arm (107mm) the way I read it. I don't think it would be powerful enough to rotate and lift a camera like the Ursa.

The other robot arm, the one in the video with the chair , could very well be used for a shot exactly like you describe.

Question is: Is it feasible to work around the robot because this model has to be bolted to the ground. Once you set it down you can't move it left or right even for an inch.

My experience has been that, on a budget at least, you often have to be flexible around the location and props, rather than the other way around.

In wonder if such models can come on a mobile platform, something with wheels that can be rolled around and what its price would be. Of course, this is more than likely what a Bolt is.

I haven't received the robot but I would assume that if it has to move to and fro that it could be placed on some kind of a dolly or slider. That would introduce another dimension of automation.

I am a bit familiar with all this stuff because i worked for some period with Milo system and due to higher engineer education i invented my own CNC machine and 3D printer
The main motion control system consists of motors, drivers for this motor, motherboard and 3D software
Motors must be not stepper but servo.Otherwise you can't control their movement in 3D space. They have to be with high torque to hold the system. That is why they are VERY expensive
Next is motherboard. It is expensive too and you have to write code for stepper motor by yourself.
Next is software. It is the most expensive part.
Add to this connectors, stepper drivers, metal frame and months of programming so you will definetely finish with pretty expensive system

The arm I bought claims to be servo controlled with an arduino with software for control. There was a time when computers and software were massive expensive things. As time went on they became mass produced items and their price came down to the point where just about anyone could own one. Robots are moving into that realm. I'm not saying that the robot I purchased is suitable for much of anything. But at $220 delivered, at minimum, it's an interesting toy.

I am a bit familiar with all this stuff because i worked for some period with Milo system and due to higher engineer education i invented my own CNC machine and 3D printer
The main motion control system consists of motors, drivers for this motor, motherboard and 3D software
Motors must be not stepper but servo.Otherwise you can't control their movement in 3D space. They have to be with high torque to hold the system. That is why they are VERY expensive
Next is motherboard. It is expensive too and you have to write code for stepper motor by yourself.
Next is software. It is the most expensive part.
Add to this connectors, stepper drivers, metal frame and months of programming so you will definetely finish with pretty expensive system

I believe you. If you bootstrap like you did I would imagine it's an uphill battle. Still, major props for getting a 3d printer and cnc machine done Not many people can say that.

This would be a major effort, for a film-bot. Would be worth it maybe if you plan to create a line and sell them because the code would be reusable but for a one-off it would not make sense.

For me right now the camerablock solution is looking pretty good.

But at $220 delivered, at minimum, it's an interesting toy.

I agree. I've seen people do amazing things with a Raspberry Pi connected to different motors.

Hey Paul, sounds great and I'm totally with you!

The thing is - the robots that mainly get much cheaper are more of the "lightweight" kind, plus maybe all the different form of "human assist/companion" robots - both are (sadly) not particular helpful in lifting the kind of cameras we use.
As you guys already stated earlier... even if you get your hands on a "cheapish" used industrial robot, these things can be very heavy, need to be bolted to the floor etc. and also there's no off the shelf motion control software/controller.. and like Paul said, the software part is expensive - just slapping a few off the shelf libraries/code parts together doesn't cut it.

Well so far I have a little arduino with a motor driver and bluetooth connection and suitable motor/belt system under a Camdolly configured as a slider on a 3m alloy tube track (the Camdolly is quite heavy, you can basically put everything on it - including yourself ;o) ... and I have an Alexmos based gimbal with a bluetooth connection as well... maybe now you can guess what my idea was/is

Basics are working and looking very promising... but yeah, you don't develop the software and motor control over night